High-precision computer vision algorithms allow Smart ID Reader to recognize the data in the video stream or from scans/photos on any device, equipped with a camera. On modern processors the capture takes around 0.07 seconds.
Smart ID Reader ensures data security by performing the recognition on the device offline. The system uses only RAM of the device, does not save or transfer data via the internet to a cloud/server for processing.
After detecting the document type, Smart ID Reader recognizes its text fields, the photo and signature. Meanwhile, the system continuously monitors the capture process and warns the user about possible errors.
In order to get the user on-boarding form filled out, the user is only required to point the document to the camera of the device. The system will automatically capture the data and populate the application form.
Multilayered document analysis
Smart IDReader does not simply extract the data — the system scans the the entire document, recognizing its text and graphic fields, including the photo and signature. Moreover, the system allows to request an assessment of the reliability of its recognition results for each document field.
Scanning in low-light conditions
Smart Engines document capture software ensures high quality recognition results regardless of device capabilities and the lighting conditions. The system has minimum lightning requirements and works on any mobile device equipped with a camera.
The full recognition cycle from “showing” the document to the camera and having the results is around 1 – 1.5 seconds.
Every frame received from the camera is processed as a separate image: the software finds the document on each frame first, and then captures its text data; the calculations for each frame then get automatically accumulated, and the system provides the final result.
QUESTIONS ABOUT SMART ID READER?
Contact Our Sales Team